Aardvark AI delivers more accurate weather forecasting, study shows
NEW YORK, UNITED STATES — A new study shows the artificial intelligence (AI) system Aardvark Weather is poised to transform the field of weather forecasting by delivering accurate predictions faster and with less computing power than traditional methods.
Aardvark uses a single machine learning model to generate forecasts from satellite and sensor data, bypassing the need for complex physics models and supercomputers.
It is developed by researchers at the University of Cambridge, supported by the Alan Turing Institute, Microsoft Research, and the European Centre for Medium Range Weather Forecasts.
The Aardvark advantage
Aardvark Weather is tens of times faster than current forecasting systems and can operate on a desktop computer, using only a fraction of the data required by traditional models.
“The weather forecasting systems we all rely on have been developed over decades, but in just 18 months, we’ve been able to build something that’s competitive with the best of these systems, using just a tenth of the data on a desktop computer,” said Richard Turner, an engineer at the University of Cambridge in the United Kingdom.
Performance and adaptability
Aardvark outperforms the U.S. national Global Forecast System (GFS) on many variables and is comparable to forecasts made by the United States Weather Service, despite using only 8% of the observational data.
While its spatial resolution is lower than some current systems, Aardvark’s ability to learn from data allows it to be tailored for specific industries or regions, such as predicting temperatures for African agriculture or wind speeds for European renewable energy companies.
Democratizing forecasting
“Aardvark’s breakthrough is not just about speed, it’s about access,” said Scott Hosking from The Alan Turing Institute.
“By shifting weather prediction from supercomputers to desktop computers, we can democratize forecasting, making these powerful technologies available to developing nations and data-sparse regions around the world.”
This innovation has the potential to revolutionize weather forecasting globally, especially in areas lacking the resources for high-resolution forecasts.
“These results are just the beginning of what Aardvark can achieve,” added Anna Allen, study coauthor, from the University of Cambridge.
“This end-to-end learning approach can be easily applied to other weather forecasting problems, for example hurricanes, wildfires, and tornadoes. Beyond weather, its applications extend to broader Earth system forecasting, including air quality, ocean dynamics, and sea ice prediction.”
